The erythrocyte chloride shift (Hamburger effect)

This chapter is most relevant to Section F8(ii) from the 2017 CICM Primary Syllabus, which expects the exam candidates to be able to "describe the carbon dioxide carriage in blood including ...the chloride shift". 

In summary: 

Westen & Prange (2003) give a reasonable overview of the situation, but their article is paywalled. So is the excellent paper by Klocke (1988) which basically goes through all the steps in the chloride shift process in excellent detail. Surely if you are going to be throwing money around you may as well buy the official exam textbook. Unfortunately, Hartog Jacob Hamburger's original paper on "Anionenwanderungen in Serum und Blut" is not available, but perhaps that is for our own good.

Definition of the chloride shift:

Westen & Prange (2003) define the chloride shift as:

"the movement of chloride ions from the plasma into red blood cells as blood undergoes the transition from arterial to venous gas partial pressures" 

There is probably something more official out there, but most authors give a description which is so close to the one above that it would be meaningless to repeat them all. In short, if this ever comes up in a viva of some sort, so long as one uses the words "chloride" and "erythrocytes" in the same sentence, one should be close to half marks already. The most important points are:

• Chloride moves into erythrocytes, and bicarbonate moves out, in venous blood.
• The reverse events take place in the pulmonary capillaries

Mechanism of the chloride shift

The molecular mechanisms for the chloride shift are described in detail below. In summary, this phenomenon is only possible because of the presence of carbonic anhydrase in RBCs. It is seen as a critically important element (as it is concentrated there, but essentially absent from the bloodstream otherwise). Without it, the reaction converting CO₂ to HCO₃^- would be painfully slow. With massive amounts of erythcyte carbonic anhydrase, we can instead count on these molecular transactions to be complete in the space of one circulatory time. In fact, because all the required proteins are available in massive concentrations, the reaction is incredibly fast. Wieth & Brahm (1980) had determined that 99% of the chloride shift process is complete within about 700 milliseconds.

In summary:

• In the peripheral capillary and venous blood:
  • CO₂ diffuses into the red cells. When the partial pressure CO₂ increases in the peripheral capillary blood due to cellular respiration, it enters the red cells fairly easily (as it is lipid-soluble). Klocke (1988) mentions offhand that its diffusion is slowed perhaps 60% by the increased viscosity of the red cell cytosol, but this is not a massive problem because the diffusion distance is about one micrometre.
  • CO₂ is converted into bicarbonate. Here, in the cytosol,  the deoxygenated haemoglobin has been acting as a proton-accepting buffer, which has increased the pH of the cytosol. The increased pH facilitates the conversion of CO₂ into bicarbonate by carbonic anhydrase. The protons produced by this process are buffered by intracellular phosphates and proteins (again, mainly deoxygenated haemoglobin)
  • The bicarbonate is exchanged for chloride by the Band 3 exchange protein, i.e. bicarbonate is removed and chloride is shuttled into the erythrocyte to maintain a neutral electrical charge. "Band 3" is the super-imaginative name given to the bicarbonate-chloride exchanger by Fairbanks et al (1971), for whom it was the third protein band from the top in the gel electrophoresis of red cell membranes. It would have been quite a fat band, as Band 3 accounts for about 25% of the total RBC membrane protein content, with over one million transport sites per cell. If it were not for the presence of haemoglobin, RBCs could easily be mistaken for a cell type responsible mainly for carrying chloride.
• In the pulmonary capillaries and arterial blood:
  • Oxygen binds to haemoglobin and causes it to release protons, i.e. decreases its buffering capacity
  • The fall in RBC cytosolic pH results in the reverse conversion of bicarbonate into CO₂ and water
  • CO₂ is then removed from the reaction by alveolar ventilation
  • As the concentration of bicarbonate in the cell falls, more bicarbonate is exchanged with chloride by the Band 3 protein.
  • Thus, there is a net decrease of bicarbonate in the blood, and a net increase in chloride

This whole thing could probably be represented better with some cartoony pictures. 

Yes, those potato-looking things are erythrocytes. The numbers came from Western & Prange (2003), whose experiments are discussed below.

The role of the Band 3 exchanger in the chloride shift

At this stage it is important to point out that the movements of chloride are a passive process.  Deranged Physiology owes a debt of gratitude to Niels Fogh-Andersen, whose work has informed the understanding of this process, and whose comments have improved the quality of this chapter.  In short, Band 3 is not an active transporter, in the sense that it uses no ATP, but a facilitated transport carrier that mediates a passive shift of chloride in (and bicarbonate out) of the erythrocyte, a process that requires no additional energy investment. The main driving forces for the action of this protein are the familiar electrochemical equilibria of chloride and bicarbonate, which are the main ions to which the erythrocyte membrane is permeable (thanks to Band 3).   In fact theoretically this system could work without Band 3, and passive diffusion was originally thought to be behind the chloride shift, but (as some early investigators pointed out) in that scenario the diffusion and equilibration of the chloride and bicarbonate would be painfully slow, as the permeability coefficient of the cell membrane to these ions is very low. All this is explained in much more professional language by Hamasaki (1999).

Magnitude of the chloride shift

With all this talk of shifting, how much chloride actually shifts? This effect is not exactly seismic. For instance, after determining what electrolyte movements should occur using quantitative physicochemical analysis, Western & Prange (2003) drained blood from healthy volunteers and subjected it to "venous-ification" by exposure to a hypoxic and hypercapnic atmosphere. At a simulated venous gas concentration, the average chloride shift of the samples was approximately 2.4 mmol/L. With a higher haematocrit, closer to 0.55 (they cheated by centrifuge but there really are people out there with such haematocrit values) the investigators were able to measure a chloride shift of around 4.3 mmol/L. 

Significance of the chloride shift:

Why is this phenomenon important? Well:

• Mitigation of pH change in the peripheral circulation: pH of the peripheral blood would change significantly more if deoxygenated RBCs were not there to buffer the acid and sequester the chloride. Westen & Prange (2003) suggest, on the basis of physicochemical modelling, that the pH of the venous blood would end up being 7.22 instead of 7.35
• Increase in the CO₂ carrying capacity of the blood:  the effect of shuttling chloride into the red cells and bicarbonate out of them increases the total potential bicarbonate carriage by the venous blood, which is good because most CO₂ is carried as bicarbonate.
• Liberation of O₂: just as CO₂, chloride is an allosteric modulator of the haemoglobin molecule. Chloride binding to the haemoglobin molecule stabilises it in the T-state, making oxygen available to the tissues. In humans, this role is probably not dominant, but in other animals it may actually be the main mediator of oxygen loading and unloading. Brix et al (1990) found that the brown bear (Ursus arctos) the chloride shift was massive (a total difference of 33 mmol/L), accounting for 40% of the total oxygen unloading in the peripheral circulation, i.e. it is the dominant modulator of the oxygen-haemoglobin association relationship.